
Prompt To App Generation
Bolt.new · Bolt
AI App Builder — Bolt (StackBlitz)
Evaluates Bolt's Prompt-to-App Generation across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI App Builder eval coverage.
About Bolt
Bolt is StackBlitz's AI app builder at bolt.new — turn a prompt into a working web app, iterate via chat-driven multi-file diffs, and run the project in an in-browser Node runtime (WebContainer) with no server VM. Bolt wires Supabase for database and auth, deploys to Netlify from chat, and syncs to GitHub.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | User opens bolt.new and types: 'Build a marketing site with a blog and SSR; I want it deployed to Netlify.' Bolt must pick a starter framework that fits SSR + Netlify-friendly output. | Pick a framework whose default build output is Netlify-deployable without extra config (e.g., Next.js, Astro, SvelteKit, Remix). State the choice in chat with one sentence of rationale before scaffolding so the user can correct on turn 1 without burning more tokens. Do not pick a CSR-only Vite + Re… | Pass / FailAi Platformhigh |
| 02 | First-turn scaffold for a Next.js app. The chat declares it will create app/, components/, lib/, package.json, next.config.js, but the diff only contains app/ and package.json. | Every file the chat declares it will create must appear in the applied diff. The WebContainer preview must boot — missing next.config.js or tailwind.config.js when the prompt asked for Tailwind is a scaffold failure. If a file is intentionally deferred, say so explicitly in chat. | Pass / FailAi Platformcritical |
| 03 | Initial scaffold writes package.json with 'react': '*' and 'next': 'latest' for the user's Next.js app. | Pin to exact or caret-pinned versions known to work together in WebContainer (e.g., 'next': '^14.2.0', 'react': '^18.3.0'). '*' and 'latest' are reproducibility hazards — every fresh install may resolve to a different working tree. Surface the pinned versions in chat so the user knows what was chos… | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Bolt
- Ai Platform
- Prompt To App Generation
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Prompt To App Generation eval for Bolt Bolt.new test?+
Evaluates Bolt's Prompt-to-App Generation across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's AI App Builder eval coverage.
How is the Prompt To App Generation eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Prompt To App Generation pack for Bolt Bolt.new contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Prompt To App Generation pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.